Photovoltaic Power Prediction of Meteorological Based on Data Drive and Mechanism Model

The proportion of photovoltaics (PV) in the grid has increased significantly with the extensive use of renewable energy. However, natural factors such as weather have a great influence on the output power of PV. Affected by the randomness and volatility of PV output power, there are unpredictable power disturbances, which can easily cause a mismatch between PV output and load demand, resulting in bus voltage fluctuations, thereby affecting the safety and stability of the system. Traditional PV power prediction models usually directly uses the data characteristics such as horizontal solar radiation flux and temperature in the meteorological forecast, ignoring the effect of solar angle on power prediction. In order to further improve the accuracy of PV power prediction, a prediction method based on the meteorological data driven and mechanism integration is proposed. Firstly, the influence mechanism of solar angle on power prediction is analyzed using the solar radiation bevel conversion principle, and the solar angle solving model is established. Secondly, the meteorological forecast data is expanded on the basis of the solar angle mechanism, the point forecast and the trend forecast models are established, and the prediction results are fused with the error index. Finally, through the test of different prediction time standards for 1 day, 3 days, and 7 days of the data of multiple real light fields, the effectiveness of the method proposed in this article was verified.